README.md

metadata

dataset_info:
  features:
    - name: pdb_id
      dtype: string
    - name: labels
      dtype: float64
    - name: SeqA
      dtype: string
    - name: SeqB
      dtype: string
    - name: source
      dtype: string
    - name: kd
      dtype: float64
  splits:
    - name: train
      num_bytes: 5998837
      num_examples: 11076
  download_size: 1697045
  dataset_size: 5998837
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

Affinity dataset from Bindwell

Natively fairly nonredundant via indel sequence similarity calculations.

Note:

There are two NaNs in the protein2_sequence column of the original dataset - we removed these rows.

Example use

from datasets import load_dataset

def get_affinity_data():
    data = load_dataset("Synthyra/ProteinProteinAffinity", split="train")
    data = data.remove_columns(['pdb_id', 'source', 'kd'])
    data = data.shuffle(seed=11)
    data = data.train_test_split(test_size=1000, seed=22)
    valid = data['test']
    train = data['train']
    valid = valid.train_test_split(test_size=500, seed=33)
    test = valid['test']
    valid = valid['train']
    return train, valid, test

Please cite

Please cite Bindwell's work if you use this dataset.